edit

+{
+ "cells": [],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

@@ -64,8 +64,8 @@ class DetectFace:
                 # specific face part
                 for (x, y) in shape[i:j]:
                     cv2.circle(clone, (x, y), 1, (0, 0, 255), -1)
-                print(name)
-                print(shape[i:j])
+                #print(name)
+                #print(shape[i:j])
 
                 face_parts[idx] = shape[i:j]
                 idx += 1
@@ -82,8 +82,8 @@ class DetectFace:
                 '''
             # visualize all facial landmarks with a transparent overlay
             output = face_utils.visualize_facial_landmarks(self.img, shape)
-            cv2.imshow("Image", output)
-            cv2.waitKey(0)
+            #cv2.imshow("Image", output)
+            #cv2.waitKey(0)
 
         # set the variables
         # Caution: this coordinates fits on the RESIZED image.
@@ -107,11 +107,11 @@ class DetectFace:
         cv2.fillConvexPoly(mask, pts, 1)
         mask = mask.astype(np.bool)
 
-        print(mask)
+        #print(mask)
 
         # Mask For background
         inversed_mask = np.logical_not(mask)
-        print(inversed_mask)
+        #print(inversed_mask)
 
 
         # Create a blank black image
@@ -119,7 +119,7 @@ class DetectFace:
         # Fill image with blue color(set each pixel to blue)
         blue[:] = [255, 0, 0]
         # extract background by applying inversed_mask
-        
+
         # extract right eye by applying polygon mask
         out2 = np.zeros_like(self.img)
         out2[inversed_mask] = blue[inversed_mask]
@@ -160,4 +160,9 @@ class DetectFace:
         cheek[0] = left
         cheek[1] = right
 
+        # show the particular face part
+        #cv2.imshow("left", left)
+        #cv2.imshow("right", right)
+        #cv2.waitKey(0)
+
         return cheek

@@ -3,6 +3,8 @@ import numpy as np
 from sklearn.cluster import KMeans
 import matplotlib.pyplot as plt
 from mpl_toolkits.mplot3d import Axes3D
+from skimage import io
+
 
 class DominantColors:
 
@@ -88,14 +90,19 @@ class DominantColors:
             g = colors[i][1]
             b = colors[i][2]
 
-            print(r,g,b)
+            #print(r,g,b)
             #using cv2.rectangle to plot colors
             cv2.rectangle(chart, (int(start), 0), (int(end), 50), (r,g,b), -1)
             start = end
 
         #display chart
-        plt.figure()
-        plt.axis("off")
-        plt.imshow(chart)
-        plt.show()
-        print(" ")
+        #plt.figure()
+        #plt.axis("off")
+        #plt.imshow(chart)
+        #plt.show()
+        for i in range(self.CLUSTERS):
+            colors[i][0] = int(colors[i][0])
+            colors[i][1] = int(colors[i][1])
+            colors[i][2] = int(colors[i][2])
+
+        return colors

@@ -25,11 +25,8 @@ class GetJson:
         i = 0
         j = 0
         for season in json_data:
-            print(season)
             for std in json_data[season]:
-                print(std)
                 for body_part in json_data[season][std]:
-                    print(body_part)
                     result_list[i][j].append(json_data[season][std][body_part])
                 j += 1
             i += 1

@@ -16,28 +16,47 @@ for color in lsk_rgb:
     lab = convert_color(rgb, LabColor, through_rgb_type=sRGBColor)
     lsk_lab.append([lab.lab_l, lab.lab_a, lab.lab_b])
 
+# 봄웜1(res/spring_1_0.png) dominant colors by order of histogram
+# skin, hair, eye 순서
+sw1_rgb = [[201.58, 158.42, 142.44], [47.38, 37.76, 35.96], [44.92, 39.05, 41.00]]
+sw1_lab = []
+for color in sw1_rgb:
+    rgb = sRGBColor(color[0], color[1], color[2], is_upscaled=True)
+    lab = convert_color(rgb, LabColor, through_rgb_type=sRGBColor)
+    sw1_lab.append([lab.lab_l, lab.lab_a, lab.lab_b])
+
+
+# 채연(res/chaeyeon.jpg) dominant colors by order of histogram
+# skin, hair, eye 순서
+cy_rgb = [[239.74, 211.85, 196.76], [16.02, 23.75, 39.83], [51.38, 35.24, 40.31]]
+cy_lab = []
+for color in cy_rgb:
+    rgb = sRGBColor(color[0], color[1], color[2], is_upscaled=True)
+    lab = convert_color(rgb, LabColor, through_rgb_type=sRGBColor)
+    cy_lab.append([lab.lab_l, lab.lab_a, lab.lab_b])
+
 getJson = GetJson()
 C = getJson.get_standard('res/standard.json')
-print(C)
+
 
 tone_analysis = ToneAnalysis()
 
 print("******************")
-a = [30, 20, 10] # 가중치
+a = [400, 300, 10] # 가중치
 spring = 0
 summer = 1
 fall = 2
 winter = 3
 result_prob = []
 for season in range(4):
-    result_prob.append(format(tone_analysis.probability(lsk_lab, season, C, a),".2f"))
-print("이성경")
+    result_prob.append(format(tone_analysis.probability(cy_rgb, season, C, a),".2f"))
+print("결과")
 print("봄   : ", result_prob[spring], "%")
 print("여름 : ", result_prob[summer], "%")
 print("가을 : ", result_prob[fall], "%")
 print("겨울 : ", result_prob[winter], "%")
 print("******************")
-if(tone_analysis.is_warm(lsk_lab)):
+if(tone_analysis.is_warm(cy_rgb)):
     if(result_prob[spring] >= result_prob[fall]):
         print("봄 웜톤")
     else:
@@ -50,7 +69,7 @@ else:
 
 '''
 # Set paths
-image = "res/lees.jpg"
+image = "res/chaeyeon.jpg"
 predictor = "shape_predictor_68_face_landmarks.dat"
 
 # Create an DetectFace instance
@@ -97,7 +116,7 @@ dc_re.plotHistogram()
 
 
 # hair
-hair_img = "res/lees_hair.jpg"
+hair_img = "res/chaeyeon_hair.jpg"
 img = cv2.imread(hair_img)
 resized_img = imutils.resize(img, width = 100)
 clusters = 6